stage might reduce transcription-coupled DNA
damage and accessibility of the genome to
transposons. After PMI, the somatic vegetative
cell is far more transcriptionally active than
the generative cell ( 25 ) and could accommo-
date transcription without an associated risk
to the germline. It will be important to estab-
lish whether SGT timing varies between
species and between male and female game-
tophytes. Is PMI a conserved moment of game-
tophyte genome activation? Or does the SGT
occur at different times in distinct plant
lineages?
The substantial increase in new transcripts
around PMI suggests that the SGT is driven by
gametophyte genome activation resulting in
new transcription, although the mechanisms
of this activation are unknown. It is unlikely
that the mitotic division itself is required to
activate transcription, as vegetative cell-like
development continues even when PMI is
blocked ( 26 – 28 ), and several gametophytic
mutants have been isolated that disrupt PMI
( 4 ). Our working hypothesis is that the SGT
begins immediately before PMI rather than
during this stage. Many substantial changes
have been observed around PMI, including
broad shifts in protein and RNA composition
( 29 ), transposon activity [inArabidopsis( 30 )],
and histone modifications ( 31 ). There is much
to learn about how these pathways are co-
ordinated to establish the independence of the
gametophyte generation.
The scope of haploid selection in predomi-
nantly diploid organisms has long been debated
( 6 ). Plants are generally accepted to experience
greater haploid selection than animals, in part
because they require many genes to complete
the haploid phase ( 2 – 4 ). By contrast, fully
enucleate animal sperm are viable and can
fertilize an egg ( 6 ). This distinction between
kingdoms may be more nuanced than previ-
ously thought: Many genes have haploid-biased
expression in mammalian sperm ( 32 ), which
suggests that animal sperm may have a greater
amount of heritable phenotypic variation than
often assumed. Our results demonstrate an
absence of haploid transcript accumulation
for half of the haploid phase in maize pollen,
limiting the time period that haploid selec-
tion may act in the male plant gametophyte.
The ability to measure allele-specific expres-
sion directly in haploid gametes and game-
tophytes will provide needed clarity on this
short but important life-cycle stage.REFERENCES AND NOTES- S. D. Tanksley, D. Zamir, C. M. Rick,Science 213 , 453– 455
(1981). - G. S. Khush, C. M. Rick,Genetica 38 , 74–94 (1967).
- B. Kindiger, J. B. Beckett, E. H. Coe Jr.,Genome 34 , 579– 594
(1991). - L. C. Boavidaet al.,Genetics 181 , 1369–1385 (2009).
- F. E. G. Beaudry, J. L. Rifkin, S. C. H. Barrett, S. I. Wright,
Plant Commun. 1 , 100115 (2020). - S. Immler,Annu. Rev. Ecol. Evol. Syst. 50 , 219– 236
(2019). - W. S. Armbruster, D. Gobeille Rogers,Am. J. Bot. 91 ,
1939 – 1943 (2004). - D.L.Mulcahy,Nature 249 , 491–493 (1974).
- G. Sandler, F. E. G. Beaudry, S. C. H. Barrett, S. I. Wright,
Evol. Lett. 2 , 368–377 (2018). - T. Lenormand, J. Dutheil,PLOS Biol. 3 , e63
(2005). - H. J. Clarke, T. N. Khan, K. H. M. Siddique,Euphytica 139 ,
65 – 74 (2004). - D. L. Mulcahy,Science 206 , 20–23 (1979).
- S. Hsu, Y. Huang, P. Peterson,Maydica 33 , 77–98 (1988).
- K. N. Schulz, M. M. Harrison,Nat. Rev. Genet. 20 , 221– 234
(2019). - G. K. Sidhuet al.,Proc. Natl. Acad. Sci. U.S.A. 112 ,
15982 – 15987 (2015). - P. M. A. Kianianet al.,Nat. Commun. 9 , 2370 (2018).
- B. Nelms, V. Walbot,Science 364 , 52–56 (2019).
- T. I. Gossmann, M. W. Schmid, U. Grossniklaus, K. J. Schmid,
Mol. Biol. Evol. 31 , 574–583 (2014).
19. R. Arunkumar, E. B. Josephs, R. J. Williamson, S. I. Wright,
Mol. Biol. Evol. 30 , 2475–2486 (2013).
20. A. M. Chettooret al.,Genome Biol. 15 , 414 (2014).
21. H. Jia, M. Suzuki, D. R. McCarty,WIREs Dev. Biol. 3 , 135– 145
(2014).
22. E. R. Valdivia, J. Sampedro, J. C. Lamb, S. Chopra,
D. J. Cosgrove,Plant Physiol. 143 , 1269–1281 (2007).
23.B.Harris,L.Dure3rd,Biochemistry 17 , 3250–3256 (1978).
24. N. Kim, S. Jinks-Robertson,Nat. Rev. Genet. 13 , 204– 214
(2012).
25. S. McCormick,Plant Cell 5 , 1265–1275 (1993).
26. C. Eady, K. Lindsey, D. Twell,Plant Cell 7 , 65–74 (1995).
27. J. Zhanget al.,Nat. Plants 3 , 17079 (2017).
28. B. Glöckleet al.,Development 145 , dev152645 (2018).
29. P. A. Bedinger, M. D. Edgerton,Plant Physiol. 92 , 474– 479
(1990).
30.R.K.Slotkinet al.,Cell 136 , 461–472 (2009).
31. M.Borg,F.Berger,Plant J. 83 , 177–188 (2015).
32. K. Bhutaniet al.,Science 371 , eabb1723 (2021).
ACKNOWLEDGMENTS
We thank J. Ross-Ibarra and E. Josephs for helpful discussions,
J. Dinneny for use of the SP8 confocal microscope, and
S. Liu for providing the A188 genome sequence ahead of
publication.Funding:This work was supported by National
Science Foundation award 17540974. B.N. was supported by
an NSF PGRP postdoctoral fellowship.Author contributions:
B.N. and V.W. designed experiments, discussed results,
and wrote the manuscript. B.N. conceived the project,
performed experiments, and conducted data analysis.
Competing interests:The authors declare no competing
interests.Data and materials availability:Sequencing data
and mapped allele-specific transcript counts are available
at the NCBI Gene Expression Omnibus (accession no.
GSE175916). All other data required to evaluate the conclusions
are in the main paper or the supplementary materials.SUPPLEMENTARY MATERIALS
science.org/doi/10.1126/science.abl7392
Materials and Methods
Figs. S1 to S12
Tables S1 to S6
References ( 33 – 49 )
MDAR Reproducibility Checklist5 August 2021; accepted 9 December 2021
10.1126/science.abl7392SCIENCEscience.org 28 JANUARY 2022•VOL 375 ISSUE 6579 429
Fig. 4. Widespread gametophyte genome activation at PMI.(A) Total transcripts detected per pollen precursor, by stage. Shown are trimmed means
(trim = 0.2) ± SE, estimated by bootstrapping. (B)(Left)Log 2 fold change in absolute transcript abundance between the late UM and BM stages for genes with
a mean expression level of at least one transcript per precursor. Solid black lines separate genes up- or down-regulated by twofold or more. (Right) Log 2
fold change in transcript abundance for transcripts mapping to the more- and less-abundant alleles (top and bottom, respectively), showing only genes
with a twofold or greater increase in overall transcript abundance. Up-regulated genes show an increase in transcript levels for the more-abundant
(haploid) allele only.
RESEARCH | RESEARCH ARTICLES